公开(公告)号：US20230401420A1

公开(公告)日：2023-12-14

申请号：US17838240

申请日：2022-06-12

Applicant: MediaTek Inc.

Inventor： Chih-Wen Goo ,  Pei-Kuei Tsung ,  Chih-Wei Chen ,  Mingen Shih ,  Shu-Hsin Chang ,  Po-Hua Huang ,  Ping-Yuan Tsai ,  Shih-Wei Hsieh ,  You Yu Nian

IPC: G06N3/04 ,  G06F9/30

CPC classification number: G06N3/04 ,  G06F9/3001

Abstract: A system receives a neural network model that includes asymmetric operations. Each asymmetric operation includes one or more fixed-point operands that are asymmetrically-quantized from corresponding floating-point operands. The system compiles a given asymmetric operation of the neural network model into a symmetric operation that includes a combined bias value. A compiler computes the combined bias value is a constant by merging at least zero points of input and output of the given asymmetric operation. The system then generates a symmetric neural network model including the symmetric operation for inference hardware to execute in fixed-point arithmetic.

公开(公告)号：US20210287339A1

公开(公告)日：2021-09-16

申请号：US17196957

申请日：2021-03-09

Applicant: MEDIATEK INC.

Inventor： Ming-En Shih ,  Yu-Cheng Tseng ,  Kuo-Chen Huang ,  Pei-Kuei Tsung ,  Hsin-Min Peng ,  Ping-Yuan Tsai ,  Kuo-Chiang Lo ,  Chun-Hsien Wu ,  Chih-Wei Chen ,  Cheng-Lung Jen

IPC: G06T3/40 ,  G06T5/00

Abstract: An image processing apparatus includes a super-resolution (SR) circuit and a resizer circuit. The SR circuit performs an SR operation upon a first image to generate a second image, wherein a resolution of the second image is not lower than a resolution of the first image, and the SR operation is based, at least in part, on one or more artificial intelligence (AI) models. The resizer circuit performs a resize operation upon the second image to generate a third image, wherein a resolution of the third image is not lower than the resolution of the second image, and no AI model is involved in the resize operation.

公开(公告)号：US20250077282A1

公开(公告)日：2025-03-06

申请号：US18820342

申请日：2024-08-30

Applicant: MEDIATEK INC.

Inventor： Ming-Hung Lin ,  Ming-En Shih ,  Shih-Wei Hsieh ,  Ping-Yuan Tsai ,  You-Yu Nian ,  Pei-Kuei Tsung ,  Jen-Wei Liang ,  Shu-Hsin Chang ,  En-Jui Chang ,  Chih-Wei Chen ,  Po-Hua Huang ,  Chung-Lun Huang

IPC: G06F9/50 ,  G06F17/15

Abstract: A digital compute-in-memory (DCIM) system includes a first DCIM macro. The first DCIM macro includes a first memory cell array and a first arithmetic logic unit (ALU). The first memory cell array has N rows that are configured to store weight data of a neural network in a single weight data download session, wherein N is a positive integer not smaller than two. The first ALU is configured to receive a first activation input, and perform convolution operations upon the first activation input and a single row of weight data selected from the N rows of the first memory cell array to generate first convolution outputs.

公开(公告)号：US20210287338A1

公开(公告)日：2021-09-16

申请号：US17118162

申请日：2020-12-10

Applicant: MediaTek Inc.

Inventor： Ming-En Shih ,  Ping-Yuan Tsai ,  Yu-Cheng Tseng ,  Kuo-Chen Huang ,  Kuo-Chiang Lo ,  Hsin-Min Peng ,  Chun Hsien Wu ,  Pei-Kuei Tsung ,  Tung-Chien Chen ,  Yao-Sheng Wang ,  Cheng Lung Jen ,  Chih-Wei Chen ,  Chih-Wen Goo ,  Yu-Sheng Lin ,  Tsu Jui Hsu

IPC: G06T3/40 ,  G06T7/00 ,  G06T5/00 ,  G06N3/04 ,  G06N3/08

Abstract: An image processing circuit performs super-resolution (SR) operations. The image processing circuit includes memory to store multiple parameter sets of multiple artificial intelligent (AI) models. The image processing circuit further includes an image guidance module, a parameter decision module, and an SR engine. The image guidance module operates to detect a representative feature in an image sequence including a current frame and past frames within a time window. The parameter decision module operates to adjust parameters of one or more AI models based on a measurement of the representative feature. The SR engine operates to process the current frame using the one or more AI models with the adjusted parameters to thereby generate a high-resolution image for display.

公开(公告)号：US20250077180A1

公开(公告)日：2025-03-06

申请号：US18820312

申请日：2024-08-30

Applicant: MEDIATEK INC.

Inventor： Ming-Hung Lin ,  Ming-En Shih ,  Shih-Wei Hsieh ,  Ping-Yuan Tsai ,  You-Yu Nian ,  Pei-Kuei Tsung ,  Jen-Wei Liang ,  Shu-Hsin Chang ,  En-Jui Chang ,  Chih-Wei Chen ,  Po-Hua Huang ,  Chung-Lun Huang

IPC: G06F7/505 ,  G06F7/57

Abstract: A digital compute-in-memory (DCIM) macro includes a memory cell array and an arithmetic logic unit (ALU). The memory cell array stores weight data of a neural network. The ALU receives parallel bits of a same input channel in an activation input, and generates a convolution computation output of the parallel bits and target weight data in the memory cell array.

公开(公告)号：US12062151B2

公开(公告)日：2024-08-13

申请号：US17118162

申请日：2020-12-10

Applicant: MediaTek Inc.

Inventor： Ming-En Shih ,  Ping-Yuan Tsai ,  Yu-Cheng Tseng ,  Kuo-Chen Huang ,  Kuo-Chiang Lo ,  Hsin-Min Peng ,  Chun Hsien Wu ,  Pei-Kuei Tsung ,  Tung-Chien Chen ,  Yao-Sheng Wang ,  Cheng Lung Jen ,  Chih-Wei Chen ,  Chih-Wen Goo ,  Yu-Sheng Lin ,  Tsu Jui Hsu

IPC: G06T3/4053 ,  G06F13/00 ,  G06N3/04 ,  G06N3/08 ,  G06T3/4046 ,  G06T5/70 ,  G06T7/00 ,  G09G5/391 ,  H01L21/033 ,  H04N9/31

CPC classification number: G06T3/4053 ,  G06F13/00 ,  G06N3/04 ,  G06N3/08 ,  G06T3/4046 ,  G06T5/70 ,  G06T7/0002 ,  G09G5/391 ,  H01L21/0338 ,  H04N9/3188 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30168

Abstract: An image processing circuit performs super-resolution (SR) operations. The image processing circuit includes memory to store multiple parameter sets of multiple artificial intelligence (AI) models. The image processing circuit further includes an image guidance module, a parameter decision module, and an SR engine. The image guidance module operates to detect a representative feature in an image sequence including a current frame and past frames within a time window. The parameter decision module operates to adjust parameters of one or more AI models based on a measurement of the representative feature. The SR engine operates to process the current frame using the one or more AI models with the adjusted parameters to thereby generate a high-resolution image for display.